Ple

ase

note

that

this

is a

n au

thor

-pro

duce

d P

DF

of a

n ar

ticle

acc

ept

ed fo

r pu

blic

atio

n fo

llow

ing

peer

rev

iew

. The

def

initi

ve p

ub

lish

er-a

uthe

ntic

ated

ve

rsio

n is

ava

ilab

le o

n th

e pu

blis

her

Web

site

1

Marine Biology Research November 2010, Volume 6, Issue 6, pages 600 - 607 http://dx.doi.org/10.1080/17451000903437075 © 2010 Taylor & Francis The original publication is available at http://www.tandf.co.uk/journals/

Archimerhttp://archimer.ifremer.fr

Imposex and butyltin concentrations in snails from the lagoon of Bizerta (Northern Tunisia)*

Youssef Lahbiba, *; Sami Abidlia; Jean-Francois Chiffoleaub; Bernard Avertyb; Najoua Trigui El Menifa

a Department of Biology, Laboratory of Environment Biomonitoring (LBE), November 7 University, Faculty of Sciences of Bizerta, Tunisia b DCN/BE Department, Laboratory of Biogeochemistry, IFREMER, Nantes, France * Published in collaboration with the University of Bergen and the Institute of Marine Research, Norway, and the Marine Biological Laboratory, University of Copenhagen, Denmark *: Corresponding author : Youssef Lahbib, Tel: +216 97 27 11 39, Fax: +216 71 88 84 67, email address : lahbibyoussef@yahoo.fr

Abstract: Imposex and butyltin concentrations were assessed in Hexaplex trunculus, Conus mediterraneus, Cyclope neritea and Nassarius mutabilis from Menzel Jemil, and H. trunculus and Stramonita haemastoma from the Bizerta channel. All females of the five species exhibited the characteristic development of male sex organs (penis and/or vas deferens). In Menzel Jemil, H. trunculus and C. mediterraneus exhibit similar imposex levels and were most affected (VDS reading 3-4). The development of a bud penis or a vas deferens sequence behind the right ocular tentacle was reported in Cyclope neritea and N. mutabilis (VDS 1). The concentration of tributyltin (TBT) was similar for the four species. The monobutyltin (MBT) was only detected in C. mediterraneus while the dibutyltin (DBT) was found in all gastropods at similar proportions as TBT. In the Bizerta channel, imposex levels were similar in both H. trunculus and S. haemastoma but imposex degree was higher in H. trunculus. This finding was confirmed by higher TBT concentration recorded in female of H. trunculus. These results suggested that H. trunculus is the more suitable bioindicator for monitoring TBT pollution. Keywords: Bioindicators; gastropods; organotins; southern Mediterranean

3

Introduction

Imposex, the superimposition of male sexual characters in female gastropods, was first

reported by Blaber (1970). The effects of this deformity vary depending on the species. In

some cases, it does not impair reproduction, but in some others, it can lead to population

decline, due to sterility and reproduction failure (Bryan et al. 1986). According to Gibbs et al.

(1997), four levels of imposex sensitivity were described: (0) snails lacking the imposex

response, (I) species develop just a penis and vas deferens, (II) snails in which the oviduct

structure and function are disrupted, and (III) species with ovary transformation to testis.

Even though organotin compounds were considered the main cause of imposex (Smith 1981;

Bryan et al. 1986), recently, hypotheses about the involvement of other factors have been

suggested. Copper (Nias et al. 1993), nonylphenol (Evans et al. 2000) polychlorinated

biphenyls (PCBs) and aroclor (Maran et al. 2006; Garaventa et al. 2008) have all been

implicated. The occurrence of imposex has been extensively documented throughout the

world. To date, this deformity has been described in 195 species worldwide (Shi et al. 2005).

According to the list reported by these authors, the banded murex Hexaplex trunculus

(Linnaeus, 1758), the cone Conus mediterraneus (Hwass, 1792), the cyclope nassa Cyclope

neritea (Linnaeus, 1758) and the oyster drill Stramonita haemastoma (Linnaeus, 1758) were

listed for gastropods and, to our knowledge, no information is published on imposex in the

variable dogwhelk Nassarius mutabilis (Linnaeus, 1758). In Mediterranean prosobranchs,

imposex studies are mainly limited to populations of S. haemastoma (Spence et al. 1990;

Rilov et al. 2000; Lemghich & Benajiba 2007), Bolinus brandaris (Linnaeus, 1758) (Solé et

al. 1998; Ramon & Amor 2001; Lemghich & Benajiba 2007), N. nitidus (Jeffreys, 1867)

(Pavoni et al. 2007) and H. trunculus (Rilov et al. 2000; Garaventa et al. 2006, 2007, 2008;

Lemghich & Benajiba 2007). Only one study on imposex in C. neritea was performed

(Quintela et al. 2006). Moreover, available data on imposex in C. mediterraneus is very

4

scarce and to our knowledge, De Fur et al. (1999) published the only document reporting the

occurrence of imposex in this snail but without data on imposex level and developmental

stages. In Tunisian waters, imposex was reported only in H. trunculus and B. brandaris

(Lahbib et al. 2008; Abidli et al. 2009). No data are available at present for S. haemastoma

despite the abundance of this species in Tunisian waters.

Organotin concentrations in gastropods have been recorded in many Mediterranean

countries such as Italy (Garaventa et al. 2007; Pavoni et al. 2007), Israel (Rilov et al. 2000)

and Malta (Axiak & Sammut 1995). However, African Mediterranean coastal studies are

limited to those of Mzoughi et al. (2004), who showed the level of organotins in the sediment

and in mussels of the Lagoon of Bizerta (Tunisia), and recently Lahbib et al. (2009) recorded

organotins in the gastropod H. trunculus from Tunisian coast. No data are available on

butyltin concentrations in C. mediterraneus, C. neritea, N. Mutabilis and S. haemastoma. The

use of snails as bioindicators of tributyltin (TBT) pollution is necessary for monitoring the

marine environment, since these species are so sensitive. For this reason, the aim of the

present study is to provide data on imposex levels and butyltin concentrations in the tissue of

five common Mediterranean gastropod species and to select from them the most suitable

species for monitoring organotin contamination. These data will be very useful prior to the

implementation of an organotin ban in Tunisia and will constitute a baseline to verify its

effectiveness.

Material and methods

Sampling and biological analysis

On 15 September 2007, 60 individuals per species were collected from 2 stations in the

Bizerta lagoon (Figure 1). On the offshore bar of Menzel Jemil, a total of 240 snails belonging

to four species (Hexaplex. trunculus, Conus mediterraneus, Cyclope neritea and Nassarius

mutabilis) were collected from a mud/sand substratum of 50 m2 area and 20 cm depth. In the

5

Bizerta channel, specimens of two species, H. trunculus and S. haemastoma, were collected

from a rocky substratum of 20 m2 area and 50 cm depth. In the laboratory, individuals were

frozen and then thawed, the shell was measured to the nearest 0.01 mm and broken with a

bench vice. The soft tissues were carefully removed and the mantles were cut longitudinally

to reveal the pallial oviduct in females. (We note here that freezing and thawing increases

penis measurements in dead specimens compared to live individuals for all studied species:

personal observation.) Sexes were determined according to the presence or absence of the

capsule gland, vagina and penis. Imposex was quantified using the following indices: (1) the

imposex incidence or frequency (I% = percentage of imposex-affected females compared to

the total number of females in the sample), (2) the vas deferens sequence index (VDSI)

following the scheme described by Stroben et al. (1992) for prosobranchs and that of Lahbib

et al. (2008) updated for H. trunculus [VDSI = (sum of imposex stage values of all

females)/(total number of females)] and (3) the relative penis length index for each female

(RPLI = female penis length x 100/average male penis length). Calculation of statistical tests

was conducted using the software Sigmastat® 3.5 for Windows. The paired sample t-test was

used to compare differences in RPLI. However, the Chisquare and Mann-Whitney rank sum

tests were used, respectively, for comparison of I(%) and VDSI.

Organotin analysis

Twenty females were selected for each species except for H. trunculus and S. haemastoma (5

females each). The operculum was removed and the soft body was finely ground using a T 18

basic Ultra-Turrax® disperser at 6000 rotations min-1. Thereafter, the tissue was freeze-dried,

weighed and maintained at -20°C in the dark until analysis. Approximately 200 mg of freeze-

dried ground tissue was used to quantify tributyltin (TBT), dibutyltin (DBT) and

monobutyltin (MBT) concentrations. The gas chromatography flame photometric detection

(GC-FPD) was used which was described in detail in Lahbib et al. (2009). The detection

6

limits were 0.8 ng Sn g-1dw for MBT, 0.7 ng Sn g-1dw for DBT and 1.0 ng Sn g-1dw for TBT.

Analysis of a certified reference material (mussel tissue BCR 477, 6 replicates) using this

procedure resulted in the following concentrations (as µg Sn g-1 dw): 1.03±0.04 for MBT,

0.75±0.03 for DBT and 0.86±0.04 for TBT. The certified values are 1.01±0.19, 0.79±0.06 and

0.90±0.08, for MBT, DBT and TBT, respectively.

Results

Imposex expression

In Cyclope neritea and Nassarius mutabilis, only one stage of imposex was revealed, the

VDS1a and VDS1b, respectively (Figures 2A-C). The first stage characterized by the

development of a bud penis behind the right ocular tentacle while the second stage showed a

sequence of the vas deferens instead of the penis. In Conus mediterraneus six stages of

imposex were revealed (1b, 1c, 2b, 3b, 3c and 4, Figure 3). The vas deferens sequence

appeared anterior to the penis, behind the right ocular tentacle (1b) or at the vaginal opening

(1c). Thereafter, these stages evolve toward 3b (a full vas deferens from the penis site to the

vagina) or 3c (a small penis and two vas deferens sequences) passing through stage 2b (a

longer vas deferens sequence behind the right ocular tentacle) to reach stage 4 (Figure 2D). In

Stramonita haemastoma, the penis appeared first (1a) followed by the vas deferens sequence

which took place at the vaginal opening (2c) or in front of the vagina (2d’). This stage then

evolves toward (3c) or 3d (a small penis attached to a vas deferens sequence) to reach stage 4

(a penis and a full vas deferens) (Figure 4). In Hexaplex trunculus, the vas deferens sequence

appeared anterior to the penis and halfway between the penis site and the vagina (VDS1d) and

evolves toward 2d (a longer vas deferens sequence) or 2d’ followed by 3b or 3d and then 4+

(Figure 4). In males, we have observed some malformations affecting the penis only in C.

neritea, namely two specimens with small penis and two were aphallic (without penis) among

the 15 analysed (Figure 5).

7

Imposex levels and butyltin analyses

In Menzel Jemil, I(%) and VDSI were similar between C. neritea and N. mutabilis and

between H. trunculus and C. mediterraneus (Table I, p > 0.05). Between these two pairs of

species, the VDSI was significantly different showing higher imposex degree in H. trunculus

and C. mediterraneus (p < 0.05). The RPLI values were too low varying between 0 (no penis)

in N. mutabilis and 0.6 in H. trunculus but without showing significant differences between

the four species (p > 0.05). The highest imposex stage (VDS4) was recorded only in C.

mediterraneus (Table I). The predominant stage was the VDS3b in C. mediterraneus against

stage VDS1d in H. trunculus. Based on these results we could classify the studied species

according to their imposex response as follow: C. mediterraneus > H. trunculus > C. neritea

> N. mutabilis. Chemical analyses showed low and similar TBT concentrations in females’

tissues from the four populations that agree with the low boating traffic (20 artisanal fishing

boats/year) found in the site (Table I). MBT was detected only in C. mediterraneus while

DBT was present in the four species (Table I). The determination of the total amount of

butyltins (BTs) showed that N. mutabilis has the highest recorded value, which disagrees with

imposex response recorded in this snail.

In the Bizerta Channel, imposex was highly developed in both species, which agrees

with the high boating traffic observed for the site (more than 500 commercial boats/year).

Imposex incidence was similar between the two snails (p > 0.05, Table I) but imposex degree

(VDSI and RPLI) was significantly higher in H. trunculus (p < 0.05, Table I). Among the four

stages of imposex found in S. haemastoma, the VDS1a was the predominant stage. In H.

trunculus, the three advanced stages which were recorded are 4, 4.3 and 5 with the VDS4

being the predominant stage (Table I). Sterility, which occurred at stage 5, was observed only

in H. trunculus at 4.3%. Sterile females showed a split capsule gland following the growth of

the vas deferens. Butyltin concentrations namely for DBT and TBT were also higher in H.

8

trunculus. This is in agreement with the results of imposex analysis. In both species, TBT was

the predominant butyltin with 50 and 52% of the total BTs, respectively, in H. trunculus and

S. haemastoma. DBT counted, respectively, for 42 and 38% and MBT for 7 and 10%. These

results suggested that H. trunculus is more sensitive to TBT pollution in terms of imposex

response and butyltin accumulation than S. haemastoma.

Discussion

Imposex features in Hexaplex trunculus, Conus mediterraneus, Cyclope neritea and

Stramonita haemastoma were found in this study, which further confirms the inclusion of

these gastropods in the list of imposex-species provided by Shi et al. (2005). However, the

development of a small vas deferens behind the right ocular tentacle in Nassarius mutabilis is

insufficient to include it in the list. According to Gibbs et al. (1997), many gastropods are

included to the list of imposex-species by virtue of the fact that females bearing a penis have

been found (an example of which is Urosalpinx cinerea). Thus, further female examinations

for N. mutabilis should be investigated at other sites namely those with high TBT

contamination in order to detect the presence of more advanced imposex stages. A more

thorough examination using histological techniques is needed to support our microscopic

observations of stage 1b. The initiation of female masculinization could be a natural

phenomenon since it was revealed in some species before the use of TBT (Gibbs et al. 1997;

Garaventa et al. 2006). Another hypothesis is related to the species’ sensitivity to TBT, which

varies according to the age and the dose (Gibbs et al. 1988). The observation of sterile

females in H. trunculus further confirms its belonging to the level II of masculinization

(Gibbs et al. 1997). For the case of S. haemastoma, imposex stages did not exceeded the

VDS4 and is consequently found only among species of level I. This observation suggested

that TBT concentrations recorded for this snail are not high enough to produce

masculinization to level II as reported for S. haemastoma from eastern Mediterranean waters

9

(Rilov et al. 2000). In C. mediterraneus, available data on imposex stages are lacking. It

seems that imposex in this species is similar to that of other members of the conidae family

because similar stages (1c, 3b, 3c and 4) observed in this study were also reported in C.

vexillum and C. betulinus (Shi et al. 2005). However, stages 1b and 2b were not mentioned by

these authors. The maximum masculinization observed in C. mediterraneus (VDS4)

suggested that TBT concentrations at the site are not high enough for imposex to reach

sterility. At present, we could classify this species at level I of masculinization until further

females from higher contaminated sites can be studied. In C. vexillum and C. betulinus, Shi et

al. (2005) have recorded sterility according to VDS 5a, 5b and 6a which suggested that

masculinization in C. mediterraneus could extend to sterility. Concerning C. neritea, imposex

was recorded at low degree (VDS 1) but in another study from Galicia (Spain), more

advanced stages were reported reaching VDS4 at which females are sterile (Quintela et al.

2006). This allows classifying C. neritea among species of level II.

The co-occurrence of gastropods in the same substratum has allowed us to

demonstrate an interspecific comparison. In Menzel Jemil, H. trunculus and C. mediterraneus

exhibited a greater incidence and intensity of imposex compared to C. neritea and N.

mutabilis. The development of the penis at an early imposex stage in H. trunculus (VDS 2d’

and 3d against VDS 4 in C. mediterraneus) suggested that this snail is more suitable for

imposex monitoring. In the Bizerta Channel, imposex response was clearly higher in H.

trunculus than in S. haematoma being in agreement with data reported by Lemghich &

Bennajiba (2007), who showed that murex species (H. trunculus and B. brandaris) are more

sensitive than Thais species (S. haemastoma). However, all these suggestions on interspecific

comparison should be supported by extending the study at other sites. In other studies,

imposex interspecific differences have also been observed between Dicathais orbita (Gmelin,

1791) and Morula marginalba (Blainville, 1832) collected from Sydney (Wilson et al. 1993),

10

between Thais clavigera (Kuster, 1852), Chicoreus capucinus (Lamarck, 1822) and Thais

gradata (Jonas, 1846) from Singapore (Tan 1999) and between H. trunculus and B. brandaris

from Tunisia (Abidli et al. 2009).

The present study reports data on imposex and TBT in some of the less-studied

gastropod species including imposex occurrence for the first time in one new Mediterranean

species in the North African coastal environment. These types of studies provide evidence

that TBT is still one of the most significant pollutants in the marine ecosystem today.

Restrictions on its use have been enforced for several years in many Mediterranean countries.

These findings are of interest for science and environmental management and can therefore

also act as a baseline study useful for the assessment of the future trends in TBT pollution in

the Tunisian coastal waters.

Acknowledgements

The authors would like to thank the editors and two anonymous referees for valuable

comments and suggestions, which greatly improved the manuscript.

References

Abidli S, Lahbib Y, Trigui El Menif N. 2009. Imposex and genital tract malformations in

Hexaplex trunculus and Bolinus brandaris collected in the gulf of Tunis. Bulletin of

Marine Science 85: 11-25.

Axiak V, Sammut M. 1995. Laboratory and field investigations on the effects of organotin

(TBT) on the oyster Ostrea edulis. Science of the Total Environment 171:117-20.

Blaber SJM. 1970. The occurence of a penis-like outgrowth behind the right tentacle in spent

females of Nucella lapillus (L.). Proceedings of the Malacological Society of London

39: 231–32.

11

Bryan GW, Gibbs PE, Hummerstone LG, Burt GR. 1986. The decline of the gastropod

Nucella lapillus around south-west England: evidence for the effect of tributyltin from

antifouling paints. Journal of the Marine Biological Association of the UK 66: 611–40.

De Fur PL, Crane M, Ingersoll CG, Tattersfield LJ. 1999. Endocrine disruption in

invertebrates: endocrinology, testing and assessment. Society of Environmental

Toxicology and Chemistry (SETAC), Pensacola, FL. 350 pages.

Evans SM, Kerrigan E, Palmer N. 2000. Causes of imposex in the Dogwhelk Nucella lapillus

(L.) and its use as a biological indicator of tributyltin contamination. Marine Pollution

Bulletin 40: 212–19.

Garaventa F, Centanni E, Fiorini S, Noventa S, Terlizzi A, Faimali M, Pavoni B. 2008. New

implications in the use of imposex as a suitable tool for tributyltin contamination:

experimental induction in Hexaplex trunculus (Gastropoda, Muricidae) with different

stressors. Cell Biology and Toxicology 24: 563-71.

Garaventa F, Centanni E, Pellizzato F, Faimali M, Terlizzi A, Pavoni B. 2007. Imposex and

accumulation of organotin compounds in populations of Hexaplex trunculus

(Gastropoda, Muricidae) from the Lagoon of Venice (Italy) and Istrian Coast

(Croatia). Marine Pollution Bulletin 54: 615–22.

Garaventa F, Faimali M, Terlizzi A. 2006. Imposex in pre-pollution times. Is TBT to blame?

Marine Pollution Bulletin 52: 701-02.

Gibbs PE, Bebianno MJ, Coelho MR. 1997. Evidence of the Differential Sensitivity of

Neogastropods to Tributyltin (TBT) Pollution, with Notes on a Species (Columbella

Rustica) Lacking the Imposex Response. Environmental Technology 18: 1219 - 224.

Gibbs PE, Pascoe PL, Burt GR. 1988. Sex change in the female dog-whelk, Nucella lapillus,

induced by tributyltin from antifouling paints. Journal of the Marine Biological

Association of the UK 68: 715-31.

12

Lahbib Y, Abidli S, Chiffoleau JF, Averty B, Trigui El Menif N. 2009. First record of butyltin

body burden and imposex status in Hexaplex trunculus from the Tunisian coast.

Journal of Environmental Monitoring 11: 1253 – 258.

Lahbib Y, Abidli S, Trigui El Menif N. 2008. Imposex level and penis malformations in

Hexaplex trunculus from the Tunisian Coast. American Malacological Bulletin 24: 79-

89.

Lemghich I, Benajiba MH. 2007. Survey of imposex in prosobranchs mollusks along the

northern Mediterranean coast of Morocco. Ecological Indicators 7: 209–14.

Maran, C., Centanni, E., Pellizzato, F., Pavoni, B. 2006. Organochlorine compounds

(Polychlorinated Biphenyls and pesticides) and Polycyclic Aromatic Hydrocarbons in

populations of Hexaplex trunculus affected by imposex in the Lagoon of Venice, Italy.

Environmental Toxicology and Chemistry 24: 486–95.

Mzoughi N, Lespes T, Bravo M, Dachraoui M, Potin-Gautier M. 2005. Organotin speciation

in Bizerte lagoon (Tunisia). Science of the Total Environment 349: 211– 22.

Nias DJ, McKillup SC, Edyvane KS. 1993. Imposex in Lepsiella vinosa from southern

Australia. Marine Pollution Bulletin 26: 380–84.

Pavoni B, Centanni E, Valcanover S, Fasolato M, Ceccato S, Tagliapietra D. 2007. Imposex

levels and concentrations of organotin compounds (TBT and its metabolites) in

Nassarius nitidus from the Lagoon of Venice. Marine Pollution Bulletin 55: 505–11.

Quintela M, Couceiro L, Ruiz JM, Barreiro R. 2006. Discovery of imposex in the gastropod

Cyclope neritea now invading Galicia (north-west Spain). Journal of the Marine

Biological Association of the UK 86:1171-1173.

Ramón M, Amor MJ. 2001. Increasing imposex in populations of Bolinus brandaris

(Gastropoda: Muricidae) in the north-western Mediterranean. Marine Environmental

Research 52: 463–75.

13

Rilov G, Gasith A, Evans SM, Benyahu Y. 2000. Unregulated use of TBT-based antifouling

paints in Israel (eastern Mediterranean): high contamination and imposex levels in two

species of marine gastropods. Marine Ecology Progress Series 192: 229-38.

Shi HH, Huang CJ, Zhu SX, Yu XJ, Xie XY. 2005. Generalized system of imposex and

reproductive failure in female gastropods of coastal waters of mainland China. Marine

Ecology Progress Series 304: 179-89.

Smith BS. 1981. Tributyltin compounds induce male characteristics on female mud snails

Nassarius obsoletus and Ilyanassa obsoleta. Journal of Applied Toxicology 3: 141–44.

Solé M, Morcillo Y, Porte C. 1998. Imposex in the commercial snail Bolinus brandaris in the

northwestern Mediterranean. Environmental Pollution 99: 241–46.

Spence SK, Hawkins SJ, Santos RS. 1990. The mollusc Thais haemastoma-an exhibitor of

imposex and potential biological indicator of tributyltin pollution. Marine Ecology 11:

147-56.

Stroben E, Oehlmann J, Fioroni P. 1992. The morphological expression of imposex in Hinia

reticulata, (Gastropoda:Buccinidae): a potential indicator of tributyltin pollution.

Marine Biology 113: 625-36.

Tan K S. 1999. Imposex in Thais gradata and Chicoreus capucinus (Mollusca,

Neogastropoda, Muricidae) from the Straits of Johor: A case study using penis length,

area and weight as measures of imposex severity. Marine Pollution Bulletin 39: 295–

03.

Wilson SP, Ahsanullah M, Thompson GB. 1993. Imposex in neogastropods: An indicator of

tributyltin contamination in eastern Australia. Marine Pollution Bulletin 26: 44–48.

14

Table I: Results of imposex and butyltin analyses in females of the studied species. N: number of females, SL: Shell length (mm), I (%): imposex incidence, VDSI: vas deferens sequence index, RPLI: relative penis length index.

Menzel Jemil Bizerta Channel C. mediterraneus C. neritea N. mutabilis H. trunculus S. heamastoma H. trunculus N 42 45 57 46 40 47 SL (mm) 33.5±3.4 10.7±0.9 22.1±2.6 47.2±3.9 57.5±4.7 58.6±5.8 I (%) 47.6 20 10,3 47.8 95 100 VDSI 0.9±1.20 0.3±0.8 0.1±0.3 0.8±0.1 2.0±1.3 4.2±0.2 RPLI 0.1±0.7 0.5±1.2 0.0 0.6±1.9 13.6±6.0 44.9±23.8 Percentage of VDS stages VDS0 52.4 80.0 89.5 52.2 5 0 VDS1a 0 20.0 0 0 40 0 VDS1b 11.9 0 10.5 0 0 0 VDS1c 11.9 0 0 0 0 0 VDS1d 0 0 0 19.6 0 0 VDS2b 2.4 0 0 0 0 0 VDS2c 0 0 0 0 27.5 0 VDS2d 0 0 0 13.0 0 0 VDS2d' 0 0 0 8.7 2.5 0 VDS3b 16.6 0 0 4.3 0 0 VDS3c 2.4 0 0 0 0 0 VDS3d 0 0 0 2.2 0 0 VDS4 2.4 0 0 0 25 59.6 VDS4.3 0 0 0 0 0 36.2 VDS5 0 0 0 0 0 4.2 Mean and standard deviation (n = 2) of butyltin concentrations (ng Sn g-1 dw) MBT 1.8±0.3 <0.8 <0.8 <0.8 3.0±0.5 3.5±0.5 DBT 2.1±0.2 1.4±0.1 2.5±0.1 2.3±0.4 11.0±2.5 20.0±0.6 TBT 2.1±0.1 2.8±0.2 3.7±0.4 1.8±0.0 15.1±1.4 23.9±0.6 Total BTs 5.9±0.0 5.0±0.3 7.0±0.5 4.9±0.4 29.1±4.3 47.4±1.6

15

Figure legends Figure 1: Sampling stations in the Bizerta lagoon. MJ: Menzel Jemil, BC: Bizerte Channel. Figure 2: Imposex stage 1a, 1b and 4 respectively in Cyclope neritea (A and B), Nassarius mutabilis (C) and Conus mediterraneus (D). Bp: bud penis, Rt: right ocular tentacle, Ec: egg capsule seen through the capsule gland tissue, Vd: vas deferens, vds: vas deferens sequence, vo: vaginal opening. Scale bar: 1mm. Figure 3 : Imposex pathways observed in Conus mediterraneus. p: penis, vo: vaginal opening, Vds: vas deferens sequence, Vd: vas deferens. Figure 4: Imposex pathways observed in Hexaplex trunculus and Stramonita haemastoma. p: penis, vo: vaginal opening, Vds: vas deferens sequence, Vd: vas deferens. Figure 5: Malformations affecting genital tract in male Cyclope neritea. A: Normal male, B: male with small filamentous penis, C: male with bud penis and incomplete vas deferens, D: aphallic male. (Rt: right ocular tentacle, P: penis, Vd: vas deferens, Lt: left ocular tentacle). Scale bar: 1mm.

16

Figure 1

Figure 2

5 km

BC

.

MJ

B izerte lagoon

N

Bizerta lagoon

Mediterranean Sea

Rt

Vds

Vds

C

Rt

Vo

Vd

D

Vd

Bp

A

Rt Bp

Ec B

Bp

Rt

17

Figure 3

vds

vds

vd p

0

1b 1c

2b

3b 3c

4

p

vo

18

Figure 4

vo

0

p

1a

vds

2c

vds

2d’

vds

3c 3d

4

1d

vds

2d

vds

3b

vd

H. trunculus S. haemastoma

19

Figure 5

D

Rt

Vd

P

Rt A

B

P

Rt

C

P

Vd

Lt